Learning View Graphs for Robot Navigation

Matthias O. Franz, Bernhard Schölkopf, Hanspeter A. Mallot, Heinrich Bulthoff

Research output: Contribution to journalArticle

140 Citations (Scopus)

Abstract

We present a purely vision-based scheme for learning a topological representation of an open environment. The system represents selected places by local views of the surrounding scene, and finds traversable paths between them. The set of recorded views and their connections are combined into a graph model of the environment. To navigate between views connected in the graph, we employ a homing strategy inspired by findings of insect ethology. In robot experiments, we demonstrate that complex visual exploration and navigation tasks can thus be performed without using metric information.

Original languageEnglish
Pages (from-to)111-125
Number of pages15
JournalAutonomous Robots
Volume5
Issue number1
Publication statusPublished - 1998 Dec 1
Externally publishedYes

Fingerprint

Navigation
Robots
Experiments

Keywords

  • Cognitive maps
  • Environment modeling
  • Exploration
  • Mobile robots
  • Omnidirectional sensor
  • Topological maps
  • Visual navigation

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence

Cite this

Franz, M. O., Schölkopf, B., Mallot, H. A., & Bulthoff, H. (1998). Learning View Graphs for Robot Navigation. Autonomous Robots, 5(1), 111-125.

Learning View Graphs for Robot Navigation. / Franz, Matthias O.; Schölkopf, Bernhard; Mallot, Hanspeter A.; Bulthoff, Heinrich.

In: Autonomous Robots, Vol. 5, No. 1, 01.12.1998, p. 111-125.

Research output: Contribution to journalArticle

Franz, MO, Schölkopf, B, Mallot, HA & Bulthoff, H 1998, 'Learning View Graphs for Robot Navigation', Autonomous Robots, vol. 5, no. 1, pp. 111-125.
Franz MO, Schölkopf B, Mallot HA, Bulthoff H. Learning View Graphs for Robot Navigation. Autonomous Robots. 1998 Dec 1;5(1):111-125.
Franz, Matthias O. ; Schölkopf, Bernhard ; Mallot, Hanspeter A. ; Bulthoff, Heinrich. / Learning View Graphs for Robot Navigation. In: Autonomous Robots. 1998 ; Vol. 5, No. 1. pp. 111-125.
@article{3287671d6c414f0f9a1d62164e9fec9a,
title = "Learning View Graphs for Robot Navigation",
abstract = "We present a purely vision-based scheme for learning a topological representation of an open environment. The system represents selected places by local views of the surrounding scene, and finds traversable paths between them. The set of recorded views and their connections are combined into a graph model of the environment. To navigate between views connected in the graph, we employ a homing strategy inspired by findings of insect ethology. In robot experiments, we demonstrate that complex visual exploration and navigation tasks can thus be performed without using metric information.",
keywords = "Cognitive maps, Environment modeling, Exploration, Mobile robots, Omnidirectional sensor, Topological maps, Visual navigation",
author = "Franz, {Matthias O.} and Bernhard Sch{\"o}lkopf and Mallot, {Hanspeter A.} and Heinrich Bulthoff",
year = "1998",
month = "12",
day = "1",
language = "English",
volume = "5",
pages = "111--125",
journal = "Autonomous Robots",
issn = "0929-5593",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Learning View Graphs for Robot Navigation

AU - Franz, Matthias O.

AU - Schölkopf, Bernhard

AU - Mallot, Hanspeter A.

AU - Bulthoff, Heinrich

PY - 1998/12/1

Y1 - 1998/12/1

N2 - We present a purely vision-based scheme for learning a topological representation of an open environment. The system represents selected places by local views of the surrounding scene, and finds traversable paths between them. The set of recorded views and their connections are combined into a graph model of the environment. To navigate between views connected in the graph, we employ a homing strategy inspired by findings of insect ethology. In robot experiments, we demonstrate that complex visual exploration and navigation tasks can thus be performed without using metric information.

AB - We present a purely vision-based scheme for learning a topological representation of an open environment. The system represents selected places by local views of the surrounding scene, and finds traversable paths between them. The set of recorded views and their connections are combined into a graph model of the environment. To navigate between views connected in the graph, we employ a homing strategy inspired by findings of insect ethology. In robot experiments, we demonstrate that complex visual exploration and navigation tasks can thus be performed without using metric information.

KW - Cognitive maps

KW - Environment modeling

KW - Exploration

KW - Mobile robots

KW - Omnidirectional sensor

KW - Topological maps

KW - Visual navigation

UR - http://www.scopus.com/inward/record.url?scp=0032024148&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032024148&partnerID=8YFLogxK

M3 - Article

VL - 5

SP - 111

EP - 125

JO - Autonomous Robots

JF - Autonomous Robots

SN - 0929-5593

IS - 1

ER -